When to prefer pair programming over peer reviews?

First lets get some help from wikipedia in defining what pair programming and peer review process is, according to wikipedia:

Now that definitions are out of the way lets talk about discussion types, there are some developers who really think only coding is work and discussions are waste of time, but for sanity’s sake lets not focus on them. There are some who value quality discussions but are skeptical of (rightly so) discussions for the sake of discussion. Or fighting over popular Tabs vs Spaces, Emacs vs Vim kind of discussion apart from those ones, I think right amount of communication is really critical between developers of same team and across teams.

I think when to discuss something is as important as what. For example there are moments during implementation when five minutes of discussion with a potential reviewer can be difference between shit hitting the fan and a pleasant smooth feature done.

But finding right balance between amount of discussions and implementation time is hard. Every developer to some extent likes tasks which they can do alone without involving other people. I think partly because open space office stuff already has lots of distractions and people finding un-interrupted time in office is becoming increasingly difficult nowadays. Also idea of having to goto a specific meeting room to discuss. Which could feel like an overkill for smaller tasks and dreadful for bigger tasks which can take hours some times.

Our team is structured in a way that every dev is kinda full-stack and supposedly can work on Android, iOS, Backend and OPS stuff. In practice most people can work on backend and ops task but there are dedicated people for Android and iOS.

If you find your team in similar situation then try to increase pairing over peer reviews. We had a decision that for each story two devs will take lead and responsibility and they can mutually decide what tasks could be done better via pairing and what is better to let someone work alone and then the other lead of the story can review it. Till now it’s working a lot better in our cross-platform team.

For me personally most times pairing is far better than waiting for review and starting other tickets and having to come back to it later when I’m in the middle of some other ticket. It’s challenging to find out specific scenarios where pairing works better than peer-reviews because those scenarios are different for each team. That’s why

What tweaks you had to make in review process which resulted in better dev workflows and team communication?

Related posts:

bibit tin kuning buah jumbo genjah manis jenis lsu gold fresh cangkok Varian bibit dijamin ORI Bibit yang kami jual berupa fresh cangkok yang terbukti mudah tumbuh dan berbuah, praktis untuk dikirim…

As we know that the latest technology signals the demise of the book. The fear will persisted throughout with the inventions of all technology that they would replace books. There is a frequently…

La prima storiella illustrata del cuoco zen Zero. “Zero si presenta” is published by Gilberto Cavallina.